Beyond ResNet: A Custom CNN Architecture for High-Performance CIFAR-10 Classification

Beyond ResNet: A Custom CNN Architecture for High-Performance CIFAR-10 Classification | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Beyond ResNet: A Custom CNN Architecture for High-Performance CIFAR-10 Classification Nnaemeka Kingsley Ugwumba This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8187931/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This paper introduces a custom Convolutional Neural Network (CNN) architecture that achieves superior performance on the CIFAR-10 dataset compared to a standard ResNet-18 baseline. Through a methodical, step-by-step enhancement process, we demonstrate that a carefully designed and optimized classical CNN can outperform a more complex, modern architecture. My baseline model achieved a strong accuracy of 88.36%. I then systematically integrated and evaluated two key enhancements: channel attention mechanisms and stochastic depth regularization. The incorporation of stochastic depth proved particularly effective, yielding our best model which attained 89.90% accuracy; a significant improvement over the 80.55% accuracy achieved by a comparably tuned ResNet-18. This research challenges the automatic preference for very deep architectures with residual connections for standard benchmarks and underscores the potential of methodically refined custom designs. Artificial Intelligence and Machine Learning Convolutional Neural Networks CIFAR-10 Systematic Improvement Attention Mechanisms Stochastic Depth Data Augmentation Deep Learning Computer Vision Experimental Methodology Model Optimization Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8187931","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":549518541,"identity":"98e7224a-f1fd-423f-9628-038c13b0e788","order_by":0,"name":"Nnaemeka Kingsley Ugwumba","email":"data:image/png;base64,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","orcid":"https://orcid.org/0009-0000-2493-9846","institution":"University of Port Harcourt, Port Harcourt, Nigeria","correspondingAuthor":true,"prefix":"","firstName":"Nnaemeka","middleName":"Kingsley","lastName":"Ugwumba","suffix":""}],"badges":[],"createdAt":"2025-11-23 23:11:33","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":true,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":true},"doi":"10.21203/rs.3.rs-8187931/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8187931/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96915047,"identity":"e1c0929f-0ef2-4801-8b9c-f6a5fe75b0d8","added_by":"auto","created_at":"2025-11-27 14:06:47","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":638084,"visible":true,"origin":"","legend":"","description":"","filename":"Sim2Safepaper.docx","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1/ae078765eccfb1841fa254b6.docx"},{"id":96757718,"identity":"89ea09de-2a72-4e07-8771-f0c954daf6d1","added_by":"auto","created_at":"2025-11-25 18:16:46","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":342,"visible":true,"origin":"","legend":"","description":"","filename":"rs8187931.json","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1/edfd02a54221a488c6cc155e.json"},{"id":96757717,"identity":"9f686d47-e017-4a03-8c34-3d5bd8f0b48d","added_by":"auto","created_at":"2025-11-25 18:16:46","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":50072,"visible":true,"origin":"","legend":"","description":"","filename":"rs81879310enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1/9c2445bf0ea7f1cf98ce56db.xml"},{"id":96757721,"identity":"3e24eea0-39b7-43f3-b3c1-9d3bd512c919","added_by":"auto","created_at":"2025-11-25 18:16:46","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":105390,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1/b232974d2a0f3a5c5a526eba.png"},{"id":96914637,"identity":"8fa8f73f-25da-4aa2-a3c0-7c30461db4a8","added_by":"auto","created_at":"2025-11-27 14:06:12","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":330605,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1/449ef072f754176bb3bda445.png"},{"id":96757729,"identity":"af0373fd-9e4e-4072-9cd5-6b42fd1950ce","added_by":"auto","created_at":"2025-11-25 18:16:46","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":107876,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1/422ab2f3a7757483ef1c97e7.png"},{"id":96757719,"identity":"7d1d33df-5eb4-44e0-8f91-d59fab5238a6","added_by":"auto","created_at":"2025-11-25 18:16:46","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56950,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1/d859613df05d84c0c324781c.png"},{"id":96757722,"identity":"8c70ee3b-f615-4776-839a-abada11ee3c3","added_by":"auto","created_at":"2025-11-25 18:16:46","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28661,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1/7e5b0caf56407d6becd08ebe.png"},{"id":96757728,"identity":"8a823ecf-8448-4984-907b-65e8f073f972","added_by":"auto","created_at":"2025-11-25 18:16:46","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":140568,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1/a955aa2b3bf18bc30b0858f2.png"},{"id":96914398,"identity":"df6bb06b-e452-4c42-9dc9-74d4a7f76a2e","added_by":"auto","created_at":"2025-11-27 14:05:52","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25898,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1/839af352d7f73a392da7a348.png"},{"id":96914629,"identity":"e3be5a1c-38c8-4012-9543-45a5aa782b6e","added_by":"auto","created_at":"2025-11-27 14:06:12","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":13628,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1/4fdba6f3ab1fadfe5b01e393.png"},{"id":96757724,"identity":"6d00d788-8f7e-4422-966b-5563f5740ac9","added_by":"auto","created_at":"2025-11-25 18:16:46","extension":"xml","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":48757,"visible":true,"origin":"","legend":"","description":"","filename":"rs81879310structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1/6a85ba79460dd5cc63cbb31a.xml"},{"id":96757723,"identity":"81f8d2bd-ef34-47a5-9172-6be9374df346","added_by":"auto","created_at":"2025-11-25 18:16:46","extension":"html","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":55336,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1/3d57a1c47f832180ad11f61f.html"},{"id":96922187,"identity":"78d03c39-134e-44d0-b8de-976b14145c8b","added_by":"auto","created_at":"2025-11-27 14:18:34","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":581127,"visible":true,"origin":"","legend":"","description":"","filename":"Sim2Safepaper.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8187931/v1_covered_2f148bab-6833-4852-bb6f-9022d52be7d5.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eBeyond ResNet: A Custom CNN Architecture for High-Performance CIFAR-10 Classification\u003c/strong\u003e\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Laskenta Technologies Limited","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Convolutional Neural Networks, CIFAR-10, Systematic Improvement, Attention Mechanisms, Stochastic Depth, Data Augmentation, Deep Learning, Computer Vision, Experimental Methodology, Model Optimization","lastPublishedDoi":"10.21203/rs.3.rs-8187931/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8187931/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis paper introduces a custom Convolutional Neural Network (CNN) architecture that achieves superior performance on the CIFAR-10 dataset compared to a standard ResNet-18 baseline. Through a methodical, step-by-step enhancement process, we demonstrate that a carefully designed and optimized classical CNN can outperform a more complex, modern architecture. My baseline model achieved a strong accuracy of 88.36%. I then systematically integrated and evaluated two key enhancements: channel attention mechanisms and stochastic depth regularization. The incorporation of stochastic depth proved particularly effective, yielding our best model which attained 89.90% accuracy; a significant improvement over the 80.55% accuracy achieved by a comparably tuned ResNet-18. This research challenges the automatic preference for very deep architectures with residual connections for standard benchmarks and underscores the potential of methodically refined custom designs.\u003c/p\u003e","manuscriptTitle":"Beyond ResNet: A Custom CNN Architecture for High-Performance CIFAR-10 Classification","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-25 18:16:41","doi":"10.21203/rs.3.rs-8187931/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"2ebaabe4-3c99-4688-8619-ac5d173b08b4","owner":[],"postedDate":"November 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":58466537,"name":"Artificial Intelligence and Machine Learning"}],"tags":[],"updatedAt":"2025-11-25T18:16:42+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-25 18:16:41","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8187931","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8187931","identity":"rs-8187931","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}